A Glimpse of Hardware Design Approaches

A Glimpse of Hardware Design Approaches

DOI: 10.4018/978-1-7998-4537-9.ch002
OnDemand PDF Download:
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

In this chapter, the authors have described that in order to design the vein enhancement and feature extraction algorithm, different modules such as DSP, embedded processor, hardware accelerator, and FPGA are implemented. Further, it has also been revealed in this chapter that the performance of the vein algorithm implemented on the Nios-II and DSP processor is not considered fast though the DSP processor is designed for signal processing applications. The FPGA is an acceptable choice for researchers due to low-cost factors. The FPGA is implemented for the hardware design of the vein algorithm. However, the performance result was not fast. Furthermore, to cater to the need for better performance, innovative hardware design architecture is the need of the time. It is observed that if there are considerable calculations in the algorithm, the optimization of the algorithm with the parallel processing capabilities of hardware will be a good choice as it can mitigate the error of the calculations.
Chapter Preview
Top

Literature Review

In (Khalil et al., 2010), a finger vein biometric feature extraction and matching system is proposed. This system implemented an embedded system on the Altera Field Programmable Gate Arrays (FPGA) and it uses Nios-II Linux Real-Time Operating System (RTOS), as shown in Figure1. Different modules of thisembedded system architecture are as follows:

  • Image Acquisition Module

  • Image Pre-processing Module

  • Feature Extraction Module

  • Matching Module

The image acquisition modulecan also be connected to the cloud if required.Further, different operations performed in the image pre-processing module can be enumerated as:

  • Color to Gray Scale Conversion

  • Grayscale Median Filter

  • Image Segmentation

  • Alignment and Re-Sizing

  • Gaussian Low-Pass Filter

  • Local Dynamic Threshold

  • Binary Median Filter

  • Thinning

Figure 1.

Embedded system architecture of the finger vein biometric system proposed in (Khalil et al., 2010)

978-1-7998-4537-9.ch002.f01

Complete Chapter List

Search this Book:
Reset